A bogie stereoscopic warehouse management method and system

Abstract

This invention relates to the field of automated warehouse management technology, and discloses a bogie-based automated warehouse management method and system, including: establishing dynamic control limits for storage locations based on historical operation coordinates to perform real-time diagnosis of the positioning coordinates of new operations, thereby distinguishing between chronic evolution and acute changes in the physical structure, and triggering dynamic offset correction or structural locking processes respectively; simultaneously, real-time monitoring of the lifting motor load during picking operations to identify and avoid abnormal mechanical resistance; this invention enables the management system to gain multi-dimensional and multi-timescale perception capabilities of the physical world. The system can not only adapt to the chronic evolution of the structure, but also respond instantly to acute changes, and perceive the hidden physical risks in the operation process, thereby transforming the system's reliability from relying on the rigidity of hardware to relying on information technology for intelligent adaptation and risk management in uncertain environments.

Description

Technical Field

[0001] This invention relates to a bogie automated warehouse management method and system, belonging to the field of automated warehouse management technology. Background Technology

[0002] In large automated storage and retrieval systems, especially those used for storing and retrieving heavy materials such as railway bogies, the management system generally follows a design and operation mode based on absolute coordinate positioning. This mode abstracts the entire warehouse into a three-dimensional coordinate system defined on engineering drawings that is theoretically unchanging. The core task of the warehouse management and control system is to instruct automated equipment such as stacker cranes to move back and forth to the theoretical coordinate points preset in the database with extremely high positioning accuracy. To achieve this goal, the industry generally invests in hardware facilities such as high-rigidity rack structures, high-precision servo motors, laser calibration systems, and precision encoders. This mode does indeed ensure the reliable execution of operations in the early stages of system operation and under ideal conditions.

[0003] However, when warehouses are used for extended periods and under heavy loads, the inherent vulnerability of this management approach, which idealizes physical entities as static models, gradually becomes apparent under an unavoidable real-world challenge: the steel racking structure of the warehouse, subjected to the combined effects of uneven loading of several tons of weight and thermal expansion and contraction caused by seasonal temperature differences, inevitably undergoes cumulative physical creep and uneven settlement. This physical evolution, which is extremely difficult to detect in a single operation, leads to a permanent, minute deviation between the actual physical center of the storage location and its theoretical design coordinates in the system database. When automated equipment faithfully executes instructions to reach that now-defunct theoretical coordinate point, occasional access anomalies begin to occur. The system logs can only record these as unattributed positioning failures or sensor malfunctions, and routine maintenance personnel responses, such as system restarts or equipment recalibration, cannot resolve this fundamental physical mismatch. Ultimately, available storage locations in areas prone to failure have to be downgraded to manual operation areas or even disabled areas, thus fundamentally eroding the core value of the automated system.

[0004] Faced with this dilemma, the most direct approach to improvement in this field is to continue using the path of precise maintenance, that is, to try to combat the natural evolution of the physical world at a higher cost by adopting more expensive materials, more stable foundations, and more frequent downtime inspections and manual calibrations. This path is obviously unsustainable in terms of economy and operational efficiency, and it does not fundamentally change the design flaw of the system blindly trusting the physical world. Specifically, the existing technology has the following shortcomings: 1. It lacks the inherent ability to perceive the slow and gradual morphological changes of the warehouse's physical structure. The system cannot identify and adapt to the slowly accumulated coordinate offset caused by material creep or uneven settlement; 2. It cannot effectively distinguish physical position deviations of different natures, and cannot identify acute and sudden structural displacements caused by accidental impacts, etc., and take the correct response strategies, which poses serious safety hazards; 3. The reliability of the system relies excessively on the initial accuracy and long-term physical stability of the hardware, and does not utilize the advantages of information technology itself to adapt to the dynamically changing physical environment. Therefore, the technical problem to be solved by this invention is how to enable automated warehouse management systems to acquire the ability to keenly perceive the microscopic morphological changes of their physical carriers simply by analyzing the data flow naturally generated during their daily operations, without adding additional dedicated measurement hardware or interrupting normal operating processes, and to dynamically correct their internal digital maps accordingly, thus achieving a shift from blind execution to adaptive operation. Summary of the Invention

[0005] This invention provides a bogie automated warehouse management method and system, the main purpose of which is to solve the problem that existing warehouse management systems, which follow static coordinate models, cannot perceive and adapt to the slow evolution and acute changes in the physical form of the shelves, resulting in a decline in the long-term reliability of the system.

[0006] To achieve the above objectives, the present invention provides a bogie automated warehouse management method, comprising the following steps:

[0007] Step a: Based on the actual location coordinates of multiple successful storage and retrieval operations at the target location, calculate a numerical range consisting of a statistical center point and a definite multiple of the standard deviation, as a dynamic control limit characterizing the statistical law of chronic evolution.

[0008] Step b: Obtain the coordinates of the new single actual positioning and determine whether it falls outside the dynamic control limits;

[0009] Step c: If the judgment result is negative, merge the coordinates into the coordinate set and recalculate the dynamic physical offset based on the updated coordinate set;

[0010] Step d: If the judgment result is yes, the coordinate is identified as an acute structural mutation event, the coordinate is prevented from entering the coordinate set, and the target storage location is locked.

[0011] Step e: When retrieving goods from the target location, monitor the load current of the stacker crane's lifting motor in real time. When the load current exceeds the reference current by a certain percentage, it is determined to be abnormal mechanical resistance and the lifting is stopped.

[0012] Preferably, before performing subsequent storage and retrieval operations on the target storage location, the method further includes algebraically summing the theoretical design coordinates of the target storage location with the latest dynamic physical offset to generate an adaptive target operation coordinate, and controlling the stacker crane to perform operations based on the adaptive target operation coordinate.

[0013] Preferably, the locking action performed on the target storage location in step d includes: setting the working status of the target storage location to a state where automatic operation is prohibited; setting the storage locations that are physically adjacent to the target storage location to a state where automatic operation is prohibited; and generating and outputting early warning information indicating that an acute structural change has occurred in the area where the target storage location is located.

[0014] Preferably, in step e, after the lifting is stopped, the method further includes controlling the stacker crane to perform an action of lowering the bogie a certain distance in the opposite direction of the lifting path.

[0015] Preferably, the rule for determining whether the monitored load current exceeds the reference current by a certain proportion in step e is defined by the following formula: ,in, For at a certain point in time The real-time monitored load current value, The reference current at time point The value, It is a fluctuation coefficient determined by the statistical distribution of current fluctuations during historical normal operation.

[0016] Preferably, the method further includes a set of verification steps for confirming the validity of the successful access operation data. This verification step is performed before step b and includes: step f, acquiring vibration signals through vibration sensors deployed on the stacker crane during the physical interaction phase of the access operation; step g, extracting the peak amplitude and oscillation decay time of the vibration signals; step h, confirming the operation as a successful access operation only when the extracted peak amplitude and oscillation decay time do not exceed their respective impact thresholds calibrated by historical high-quality operation data, and using the corresponding actual positioning coordinates for the judgment in step b.

[0017] Preferably, the actual positioning coordinates in step a are three-dimensional coordinate readings provided by a laser rangefinder or encoder installed on the stacker crane at the instant the physical action of confirming the completion of the storage and retrieval operation.

[0018] Preferably, the method further includes: when the stacker is in an idle state, controlling the stacker to move to one or more absolute reference points on the warehouse floor whose three-dimensional coordinates remain constant for positioning measurement, and correcting the systematic measurement drift of the stacker's own positioning sensor based on the result of the positioning measurement.

[0019] Preferably, in step c, a dynamic physical offset is recalculated based on the updated coordinate set. Specifically, this involves: calculating the arithmetic mean of all actual positioning coordinates in the updated coordinate set on each coordinate axis in the three-dimensional coordinate system, and using this arithmetic mean as the new statistical center point; then, calculating the vector difference between the new statistical center point and the theoretical design coordinates of the target cargo location, and using this vector difference as the dynamic physical offset; the determined multiplier is an integer greater than or equal to three selected based on the risk level assessment of the method to be protected.

[0020] A bogie automated warehouse management system, the system comprising:

[0021] A coordinate history database is used to store the actual location coordinates of multiple historical successful access operations obtained for target storage locations in the warehouse;

[0022] A control limit generation unit is connected to the coordinate history database. The control limit generation unit is configured to calculate and generate a numerical range based on the coordinates in the coordinate history database. This range consists of a statistical center point and a determined multiple of the standard deviation of the coordinates plus or minus the statistical center point. The numerical range is used as a dynamic control limit to characterize the statistical law of the slow evolution of the target cargo location.

[0023] A data acquisition interface is used to obtain the single actual positioning coordinates corresponding to the new successful access job after the new access job is completed;

[0024] A central processing unit (CPU), connected to the coordinate history database and the data acquisition interface, is configured to perform the following operations: determine whether the value of the single actual positioning coordinate acquired by the data acquisition interface falls outside the dynamic control limit generated by the control limit generation unit; if the determination result is negative, aggregate the single actual positioning coordinate through the coordinate history database and recalculate a dynamic physical offset based on the updated coordinate set; if the determination result is positive, identify the single actual positioning coordinate as a characterization signal of an acute structural mutation event, prevent the coordinate from updating the dynamic physical offset, and generate an instruction to perform locking action on the target cargo location;

[0025] A motor current monitoring module is used to monitor the load current of the stacker crane's lifting motor in real time when performing the operation of retrieving goods from the target storage location; and the central processing unit is also configured to: when the load current monitored by the motor current monitoring module exceeds the reference current of the motor under the same working conditions without load by a certain proportion, determine that there is abnormal mechanical resistance and generate a command to stop lifting; a device control actuator, connected to the central processing unit, is used to receive and execute the locking disposal command and the command to stop lifting.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. By establishing statistical control limits for historical positioning data, the system can instantly analyze each new operational positioning data. This mechanism enables the system to distinguish between two completely different types of physical changes: one is the gradual and slow physical evolution of the shelving due to long-term load-bearing or temperature differences; the other is the instantaneous and significant structural positional change caused by factors such as accidental impacts. For positioning data falling within the control limits, the system treats it as a manifestation of the slow evolution of the physical form and continuously and smoothly corrects its internal digital map, so that operational instructions can dynamically fit the actual position of the material carrier. For positioning data falling outside the three control limits, the system identifies it as a signal of a structural change event, immediately stops updating the digital map and isolates the data, and simultaneously executes a locking or active detection process on the relevant area. This avoids continuous operational failures or aggravated physical damage caused by following outdated instructions during the adaptive delay window.

[0028] 2. In addition to the real-time monitoring of the load current of the lifting motor during the picking operation, this invention also introduces real-time monitoring of the load current of the lifting motor. The system continuously compares the real-time current profile during the picking and lifting process with the pre-learned and stored reference load profile corresponding to the normal resistance-free lifting process. If the real-time current profile deviates significantly from the reference profile when the stacker crane positioning system confirms that the position is correct, but the lifting action has just begun, the system will determine that there is a slight mechanical interference or soft hook state between the bogie and the rack that cannot be detected by conventional positioning sensors. At this time, the system will immediately stop the lifting action and execute a safety return procedure. This transforms an invisible dynamic process risk into a manageable digital event in the instant before physical damage or component fatigue actually occurs, thereby adding a layer of force-sensory safety assurance in the process execution to the accuracy of the main solution in spatial positioning.

[0029] 3. When confirming the completion of a storage or retrieval operation, this invention also acquires and analyzes the vibration timing characteristics of the final stage of the operation through vibration sensors deployed on the equipment. The system uses this information to determine whether a smooth soft landing was achieved between the bogie and the forks or rack, or whether a hard landing with physical impact occurred. Only when the vibration characteristics are within the preset normal range is the system recognized as a high-quality successful operation, and its corresponding final positioning coordinates are allowed to enter the statistical pool mentioned in the first point for updating the digital map. If an impact is determined to have occurred, the positioning data of that operation will be discarded or given a lower weight. This mechanism constructs a physical process-level quality gatekeeper for the core data source of the entire adaptive positioning system. It ensures that only high-quality data that truly reflects the static position of the storage location and is not contaminated by operational impacts can be used for long-term learning and correction of the system, thereby improving the purity and reliability of the system's perception of chronic deformation in the physical world.

[0030] 4. Through the coordinated operation of the above multiple mechanisms, this invention transforms the warehouse management system from an instruction execution unit that can only issue and verify the coordinates of the endpoint location into a management system with multi-dimensional and multi-time-scale perception capabilities. It can not only perceive the slow aging of physical structures through the statistical trends of location data, but also identify acute trauma through the dispersion of single data. It can not only detect the hidden resistance in the operation process through the slight changes in motor current, but also infer the process quality of physical interaction through the characteristics of structural vibration. These information flows, which originate from different physical phenomena and are obtained through different sensors, form a logical closed loop of mutual verification and mutual constraint within the system. This allows the system to make decisions that no longer rely on any single-dimensional information when dealing with the uncertainties of the real world, but are based on a more complete and profound internal understanding of physical reality. Attached Figure Description

[0031] Figure 1 This is a flowchart illustrating a bogie automated warehouse management method according to the present invention.

[0032] Figure 2 This is a sequence diagram of the interaction between key units of the management system of this invention;

[0033] Figure 3 This is a schematic diagram of the top-level process for automated warehouse inbound and outbound operations according to the present invention.

[0034] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0035] The technical solutions in the embodiments of the present invention will be clearly and completely described below. The described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0036] This application provides a bogie automated warehouse management method and system. By analyzing and utilizing daily operation data of automated equipment, the management system can adaptively perceive and manage risks related to various changes in the physical world of the warehouse. The operation process is as follows: First, the system establishes dynamic control limits for each storage location, representing normal fluctuations in its position, through statistical analysis of historical operation coordinates. Based on these limits, the system performs real-time diagnosis of the positioning data for each new operation, thereby distinguishing between chronic evolution of the physical structure caused by material creep or thermal expansion and contraction, and acute structural changes caused by factors such as accidental impacts. The system then triggers two response processes: dynamic offset correction or structural safety locking, respectively. Furthermore, the system introduces real-time monitoring of the load current of the lifting motor during the picking operation and analyzes the temporal characteristics of structural vibration during the storage and retrieval interaction stage. This is done to identify and avoid potential mechanical interference risks and ensure the effectiveness of the positioning data, ultimately forming an operation mechanism that ranges from spatial positioning to process monitoring and from chronic adaptation to acute response.

[0037] In automated storage and retrieval systems (AS / RS) that bear heavy railway bogies for extended periods, the steel racking structures typically undergo slow and cumulative physical deformation under the combined effects of continuous non-uniform loading and seasonal temperature variations. This causes the actual physical center of the storage location to deviate from its theoretical design coordinates recorded in the system database. To address this chronic, gradual physical change, the management method provided in this invention is configured to execute an adaptive mapping process based on statistical trends. This process begins by establishing and maintaining a queue in the coordinate history database for each target storage location in the warehouse, storing the actual positioning coordinates of its historical successful access operations. These actual positioning coordinates are three-dimensional coordinate readings provided by a laser rangefinder or encoder installed on a stacker crane at the moment the physical access action is confirmed. Once a preset number of historical coordinate samples have accumulated in the queue, a control limit generation unit is triggered. Based on this coordinate set, it first determines the statistical center point of the storage location by calculating the arithmetic mean of all coordinates on each axis of the three-dimensional coordinate system, and then further calculates the standard deviation of these coordinates relative to this center point. The system then performs algebraic operations on a fixed multiple of the standard deviation and the statistical center point to generate a numerical interval. This fixed multiple can be selected based on statistical principles. For example, to obtain a higher confidence level, an integer greater than or equal to three can be selected. This interval is defined as the dynamic control limit characterizing the statistical law of the chronic evolution of the target cargo location. In this way, the system establishes a statistically significant definition of the normal fluctuation range of the physical location of each cargo location.

[0038] Given that the physical form of a warehouse can undergo not only slow evolution but also significant acute structural changes, and that these two distinct types of change require the system to possess differentiated response capabilities, the central processing unit (CPU) first determines whether the coordinate value falls outside the dynamic control limits after a successful storage / retrieval operation and when the data acquisition interface obtains the single actual positioning coordinates corresponding to that operation. If the result is negative, the system classifies it as a manifestation of slow physical structural evolution and incorporates the single actual positioning coordinates into the coordinate set representing slow evolution. Based on this updated coordinate set, the CPU recalculates a new statistical center point and calculates the vector difference between this new statistical center point and the theoretical design coordinates of the target storage location. This vector difference is used as the latest dynamic physical offset. Before executing subsequent storage / retrieval operations for the target storage location, the system will... The theoretical design coordinates of the storage location are algebraically summed with the latest dynamic physical offset to generate an adaptive target operation coordinate. The stacker crane is then controlled to perform operations based on this adaptive target operation coordinate. Conversely, if the judgment result is yes, the central processing unit identifies the single actual positioning coordinate as a characterization signal of an acute structural change event and immediately executes a set of defensive procedures. These procedures first prevent the abnormal coordinate from entering the coordinate set used to characterize chronic evolution to prevent it from affecting the statistical benchmark for long-term trend analysis. At the same time, the target storage location is locked. This locking procedure may include setting the target storage location and its physically adjacent storage locations to a state where automatic operation is prohibited, and generating early warning information indicating that an acute structural change has occurred in the area. This discriminative response mechanism enables the system to adapt to slow structural aging while also instantly sensing and isolating the risks brought by acute trauma.

[0039] To address the risk that even with accurate positioning, minor mechanical interference between the bogie and the rack may still occur during the retrieval process, this invention introduces an online pre-inspection mechanism based on motor load monitoring when retrieving goods from the target location. This mechanism uses the stacker crane's lifting motor driver as a sensor capable of reflecting minute load changes. During initial system deployment or calibration, the stacker crane is pre-runned one or more lifting operations under confirmed no-load and unobstructed conditions. The motor current monitoring module records the current-time change curve of this process, using it as a reference current. Storage; During actual picking operations, when the stacker crane lifts the bogie, the motor current monitoring module monitors the load current of the lifting motor in real time. The central processing unit continuously compares the real-time current value with the reference current value at the same time point. Once it determines that the monitored load current exceeds the reference current of the motor under the same operating conditions by a certain proportion, the determination rule is met. If this occurs, it is determined that there is abnormal mechanical resistance and a command to stop lifting is generated. It is a floating coefficient determined by the statistical distribution of current fluctuations during historical normal operation; after the equipment control actuator receives the command to stop lifting, it will immediately stop the lifting action and can further control the stacker crane to perform a safe retraction action to lower the bogie a certain distance in the opposite direction of the lifting path.

[0040] Furthermore, considering that the accuracy of the entire adaptive positioning correction system depends on the quality of its input data source, and that physical impacts during operation can cause instantaneous vibrations in the stacker crane, thus affecting the final reported positioning coordinate readings, this invention also introduces a set of verification steps to confirm the validity of successful access operation data. This verification step involves deploying vibration sensors on the stacker crane to acquire vibration signals during the operation interaction phase. The central processing unit then extracts two temporal features of the vibration signal: peak amplitude and oscillation decay time. The system only confirms the operation as a high-quality successful access operation when both the extracted peak amplitude and oscillation decay time do not exceed their respective impact thresholds calibrated by historical high-quality operation data. The corresponding actual positioning coordinates are then used for the aforementioned dynamic control limit judgment and dynamic physical offset update calculation. If either feature exceeds the threshold, the positioning data for that operation is determined to be affected by an impact and will be discarded or assigned a lower weight. This constructs a physical process-level verification mechanism for the data source of its adaptive positioning system, ensuring that the information used for long-term system learning and correction is high-quality data that reflects the static position of the cargo location.

[0041] To ensure the long-term stability of the entire measurement system and avoid potential systematic measurement drift from the stacker crane's own positioning sensors, this invention also includes a self-calibration mechanism. This mechanism is achieved by pre-setting one or more absolute reference points on the warehouse floor whose three-dimensional coordinates remain constant. When the stacker crane is idle, the system controls it to move to these absolute reference points for positioning measurements. The system then compares the results of these measurements with the known precise coordinates of the reference points to calculate and correct the systematic measurement drift of the stacker crane's own positioning sensors. This ensures the accuracy of the stacker crane itself as a measuring tool, thereby guaranteeing its long-term reliability in measuring the object being measured.

[0042] Example 1: In a railway bogie automated warehouse that has been in heavy-duty operation for over two years, the southern area has experienced cumulative physical creep in some of the rack steel structures due to long-term non-uniform loading and seasonal temperature differences. This has led to frequent positioning failure alarms at multiple storage locations in this area. To maintain operational continuity, the automated operation permissions for this area have been systematically downgraded to manual intervention mode. To address this situation, the management method described in this invention is introduced. The system first utilizes successful access operation data stored in the past operations of this problematic area to generate initial dynamic control limits for each target storage location. During the initial operation phase, because some historical coordinate data contains... The deviation caused by unidentified physical impacts results in fluctuations in the calculated dynamic physical offset. Subsequently, in each subsequent storage and retrieval operation, the system uses vibration sensors deployed on the stacker crane to analyze the vibration timing characteristics during the physical interaction phase. Only when the peak amplitude and oscillation decay time of an operation do not exceed the preset impact threshold are the corresponding actual positioning coordinates confirmed as valid data and included in the coordinate set used to characterize chronic evolution. This preliminary verification step provides filtered data input for the subsequent calculation of dynamic physical offsets, enabling the offset to more stably track the actual displacement of the rack caused by physical creep.

[0043] After the method was deployed for a period of time, a maintenance vehicle accidentally and slightly impacted a rack upright in the area, causing a momentary physical displacement of a storage location connected to the upright. When the stacker crane arrived at the storage location and completed its operation in the next task, the actual positioning coordinates it reported for a single instance differed significantly from the statistical center point of the storage location calculated based on historical data, and its value fell outside the range defined by the dynamic control limits of the storage location. The system immediately determined this to be a characteristic signal of an acute structural change event and immediately implemented a locking measure. This measure prevented the abnormal coordinates from updating the dynamic physical offset and simultaneously locked the target storage location and... Physically adjacent storage locations are set to a state where automatic operation is prohibited, and a warning message indicating that a structural anomaly has occurred in the area is output to the management terminal. This response method avoids the risk of continuous operation failure or aggravated physical damage caused by following outdated instructions based on slow evolution during the adaptive delay window. After manual inspection and physical repair, the area is unlocked, and the system resumes automatic operation. It no longer relies on a static coordinate system based on the initial engineering drawings for management, but keeps its internal digital map synchronized with the physical state of the storage structure by continuously analyzing the positioning data stream generated in daily operations.

[0044] Example 2: To objectively verify the effectiveness of the management method described in this invention in dealing with the chronic evolution and acute changes in the physical structure of a warehouse, this comparative experiment was designed. The purpose was to quantify the performance difference of this method in terms of positioning success rate and abnormal event response capability compared to the traditional fixed-coordinate-based management method. This experiment was conducted on a standard industrial-grade automated testing platform, which included a stacker crane and a row of shelves. One storage location was selected as the target storage location. To simulate chronic physical deformation, before the experiment began, a distance was artificially introduced between the theoretical design coordinates of the target storage location in the warehouse management system and its actual center position in the physical world, through software settings. Axial direction , Axial direction The fixed deviation; to simulate acute structural events, a trigger is included in the test procedure, which can introduce an additional force at the target storage location at a specified time. Axial direction The instantaneous positional change; the experiment was divided into a control group and an experimental group. The control group adopted a traditional warehouse management system that follows absolute coordinate positioning, while the experimental group adopted the management method described in the specific implementation method. Both groups performed 500 consecutive inbound and outbound cycles on the target storage location under the same physical platform and deviation conditions. The number of cycles was set to provide sufficient data samples for the adaptive algorithm of the experimental group to converge to a steady state, while the total duration was controlled within a single standard operation shift. The acute event trigger point was set before the start of the 250th cycle to ensure that both variant modes were tested.

[0045] During the experiment, if the trend of the number of positioning failures with the number of operation cycles is depicted, the two groups show significant differences. The stacker crane in the control group always targets a fixed theoretical design coordinate. Due to a chronic deviation between this coordinate and the actual physical center of the storage location, it accumulated 187 positioning failures in the first 249 cycles. After an acute mutation was introduced in the 250th cycle, all subsequent 251 operations failed, and the system did not detect or respond to this mutation event. In contrast, the experimental group experienced 12 positioning failures in the early stages of operation due to the initial chronic deviation. However, as the number of operations increased, its dynamic physical offset was continuously corrected through statistical learning of the coordinates of successful operations, and the positioning failure phenomenon quickly disappeared. When an acute mutation was introduced after the 250th cycle, the experimental group experienced a positioning failure during that operation. Its reported coordinates were immediately identified by the system because they fell outside the dynamic control limits. The warehouse location was locked down, and subsequent operations were suspended.

[0046] Experimental data shows that the control group using fixed coordinates maintained a high positioning failure rate when faced with chronic coordinate shifts, and completely lacked the ability to detect acute changes, ultimately rendering the storage location completely unusable. The management method described in the experimental group, through the mechanism of dynamic physical offset, can effectively adapt to chronic coordinate shifts caused by physical creep, and through the mechanism of dynamic control limits for real-time diagnosis of single operation data, it can identify and execute safety locks in the first operation when an acute structural change occurs. This proves that the method can maintain the long-term operational reliability of the system and the safety response capability to emergencies without relying on manual intervention.

[0047] Example 3: To illustrate the implementation of a bogie automated warehouse management method and system, as follows... Figure 1As shown: The process begins after the storage and retrieval operation starts. The actual positioning coordinates of the operation are obtained using a laser rangefinder or encoder on the stacker crane. Then, the vibration signal validity verification stage is entered to determine if the operation is a high-quality, successful operation. If the result is negative (impact occurred), the positioning coordinates are considered to have data contamination and are discarded or assigned a lower weight. If the result is positive (high-quality data), the coordinates enter the core diagnostic stage, which determines whether the coordinates fall outside the dynamic control limits to distinguish between chronic and acute displacement. If the result is negative (chronic evolution), the system incorporates the valid coordinates into the coordinate set for recalculating the dynamic physical offset and ultimately generating an adaptive... The target work coordinates are used for subsequent precise positioning. If the judgment result is (acute mutation), the system will identify it as an acute structural mutation event and prevent the abnormal coordinates from entering the statistical pool. At the same time, the target storage location will be locked and an early warning message will be generated and output to notify the management personnel to conduct a manual inspection. Meanwhile, in the key module of parallel execution of the picking operation: process risk monitoring, the system will monitor the load current of the lifting motor in real time to detect hidden physical risks in the operation process. Once it is determined that there is abnormal mechanical resistance, the system will immediately stop the lifting and perform a safe retreat, thus forming a closed-loop management process from spatial positioning to process monitoring, from chronic adaptation to acute response.

[0048] like Figure 2 As shown in the diagram: The process begins with the operator issuing a storage / retrieval command to the management system. The management system then sends a request to the central processing unit (CPU) to query the coordinates of the target storage location. To generate adaptive coordinates, the CPU first requests the latest dynamic physical offset from the coordinate history database. After obtaining the offset data, the CPU calculates the adaptive target coordinates and sends a positioning control command to the stacker crane. After the stacker crane moves to the target location and performs the storage / retrieval action, it reports the actual positioning coordinates to the CPU through the data acquisition interface. After transmitting the coordinate data, the CPU sends the coordinates to the control limit generation unit for judgment on whether the coordinates are within the control limits. Upon receiving confirmation that the judgment result is yes, the CPU performs an operation to incorporate the coordinates from the coordinate history database into the chronic evolution set and triggers a recalculation of the dynamic physical offset to update the offset data. Finally, the CPU sends a job completion notification to the management system, and the management system displays the job success to the operator. This fully demonstrates the complete collaborative process of each part of the system from command reception to adaptive learning.

[0049] like Figure 3As shown in the diagram, starting from the beginning, the system is divided into two main branches: inbound and outbound operations. In the inbound branch, the system checks the storage location. If an error is found or there is no available space, an alarm is triggered, and feedback information / manual processing is initiated to the higher-level system. If the storage location is OK, the bogie data is further verified. After verification, the WCS controls the stacker crane and conveyor system to perform the bogie inbound operation. After the operation is completed, feedback inbound information is sent to the higher-level system. In the outbound branch, the process is similar to that of inbound. Its core checkpoints are checking inventory and verifying bogie data. If problems such as missing information or incorrect information occur, an alarm and manual processing will also be triggered. This diagram reveals the top-level business environment embedded in the management method described in this invention.

[0050] Example 4: To achieve precise control during bogie retrieval operations, specifically, even if the stacker crane's three-dimensional spatial positioning is accurate during the physical execution phase of the bogie retrieval operation, the non-load-bearing components of the bogie may still experience slight twisting or tolerance accumulation between themselves and the rack structure due to temperature differences. This creates a subtle mechanical interference state that conventional sensors cannot detect. When the stacker crane vertically lifts the bogie, this state may lead to transient mechanical resistance caused by long-term fatigue or wear of the components. To monitor and mitigate this risk online, it is necessary to improve the real-time monitoring mechanism of the motor load current and establish a standardized offline calibration and parameter determination procedure. This procedure is executed when the stacker crane is in an idle maintenance state. The technical object it operates on is the stacker crane's lifting motor and its servo drive, which has a speed of not less than... The ability to output real-time load current data to external systems at a specific frequency; the calibration process begins with building a reference current. The model shows an operator controlling a stacker crane using a standard-weight bogie, continuously executing operations at a confirmed location free from any physical interference. During each complete lifting operation, the motor current monitoring module records a time-stamped sequence of load current data for each operation, from start to reaching the designated height. Subsequently, the central processing unit processes this data. Align the time series data sets and target each time point in the improvement process. Calculate this moment The arithmetic mean of the current values ​​constitutes the reference current profile characterizing the normal lifting process under this operating condition. .

[0051] Furthermore, in order to determine the float coefficient used to identify anomalies... The procedures will continue to use the aforementioned data collection methods. The central processing unit (CPU) analyzes the current data at each point in time during the boosting process. Further calculation of this moment Each current value is relative to the reference current at that time point. Standard deviation The standard deviation The normal fluctuation range of the motor load current during each stage of the lifting cycle under the condition of no abnormal resistance was quantified. The logic for determining whether there is abnormal mechanical resistance lies in judging whether the deviation between the real-time current value and its expected baseline value at the current time point exceeds its normal statistical fluctuation range, and an instantaneous current threshold is used to trigger the suspension of lifting. It can be defined as: ,in, A dimensionless coefficient defined based on risk assessment is set as follows in this procedure: Define this threshold in accordance with the judgment rules given in the specific implementation: By establishing a correlation, the fluctuation coefficient can be adjusted. At different times The value is determined by the following formula: Its values ​​have a statistical basis and a defined calibration process.

[0052] After the above calibration is completed, when the system detects abnormal mechanical resistance during actual operation and stops lifting, its subsequent safe retraction action, namely controlling the stacker crane to lower the bogie a certain distance in the opposite direction of the lifting path, is set as an engineering value. This value is set to be larger than the typical geometry of structural features that may form hooks on bogies or racks in order to achieve effective separation.

[0053] Example 5: To determine the vibration timing characteristic benchmark for differentiating work quality, a preliminary impact threshold calibration procedure is performed when the system is initially deployed or a new bogie model is added for access. This procedure is conducted in an environment where external interference is confirmed to be absent. The operator sets the system to a calibration mode and selects the target bogie model as the calibration load. In this mode, the stacker crane applies a gentler parameter to the bogie at a speed lower than the normal operating speed. These complete access cycles, performed under controlled conditions, are used as samples to define high-quality jobs.

[0054] During the aforementioned calibration process, vibration sensors deployed on the stacker crane collect and record vibration signals at each physical interaction stage, thereby obtaining... A set of independent vibration time-series characteristic data, each set containing a peak amplitude reading and an oscillation decay time reading; after the calibration process is completed, the central processing unit first calculates these... Arithmetic mean of peak amplitude with standard deviation ,as well as Arithmetic mean of the group oscillation decay time with standard deviation Subsequently, based on these statistical results, the system determines the final impact threshold using a preset threshold calculation rule, where the impact threshold for peak amplitude is set as follows: The impact threshold for oscillation decay time is set to These calculated thresholds are then stored in the system and associated with the corresponding bogie model, serving as the basis for determining whether an impact exists during subsequent routine operations.

[0055] Example 6: When the management system is initially deployed in a newly built bogie automated warehouse, in order for its core diagnostic mechanism, which distinguishes between chronic evolution and acute mutation of physical structure, to take effect immediately without the accumulation of historical operational data, a standardized initial baseline model construction procedure needs to be implemented.

[0056] This procedure places the system in an initial data acquisition mode. In this mode, the system controls the stacker crane to move to each storage location in the warehouse according to a preset path while unloaded. It then uses its onboard laser rangefinder to perform a high-precision positioning measurement of the physical center of each storage location. The resulting set of coordinate readings covering all storage locations in the warehouse is used as seed data for constructing the initial baseline model. Based on this, the system directly sets the coordinates of the single high-precision measurement for each storage location as the initial statistical center point for that location. Simultaneously, the system calls upon a globally unified initial standard deviation determined based on the stacker crane's own designed positioning accuracy and repeatability accuracy indicators. This value reflects the inherent fluctuation range of the equipment's own positioning readings; subsequently, the system calculates an initial dynamic control limit for each storage location, the range of which is the initial statistical center point of that storage location plus or minus the initial standard deviation. A certain multiple constitutes the initial dynamic control limit setting for all storage locations. Once the initial dynamic control limit setting for all storage locations is completed, the system can enter the normal operation state. In subsequent operation, as real successful storage and retrieval operation data accumulates, the initial statistical center point and initial standard deviation of each storage location will be replaced by new values ​​obtained from multiple actual positioning coordinate statistics.

[0057] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A bogie stereoscopic warehouse management method characterized by, The method comprises the following steps: Step a, based on the actual positioning coordinates of a plurality of successful access operations of the target storage location, a numerical interval composed of a statistical center point and a determined multiple of a standard deviation is calculated as a dynamic control limit representing the statistical law of chronic evolution; Step b, the new single actual positioning coordinate is obtained, and it is judged whether it falls outside the dynamic control limit; Step c, if the judgment result is no, the coordinate is integrated into the coordinate set, and the dynamic physical offset is recalculated based on the updated coordinate set; Step d, if the judgment result is yes, the coordinate is identified as a structural acute mutation event, the coordinate is prevented from entering the coordinate set, and a locking treatment is performed on the target storage location; Step e, when the target storage location is executed, the load current of the stacker lifting motor is monitored in real time, and when the load current exceeds the reference current by a certain percentage, it is determined that the abnormal mechanical resistance is abnormal and the lifting is stopped.

2. The bogie stereoscopic warehouse management method according to claim 1, wherein Before performing the subsequent access operation of the target storage location, the method further comprises algebraically summing the theoretical design coordinate of the target storage location and the latest dynamic physical offset to generate an adaptive target operation coordinate, and controlling the stacker to perform the operation according to the adaptive target operation coordinate.

3. The bogie stereoscopic warehouse management method according to claim 1, wherein The locking treatment performed on the target storage location in step d includes: setting the working state of the target storage location to a state in which automatic operation is prohibited; and setting the storage locations adjacent to the target storage location in physical structure to a state in which automatic operation is prohibited; and generating and outputting warning information indicating that a structural acute mutation occurs in the area where the target storage location is located.

4. The bogie stereoscopic warehouse management method according to claim 1, wherein After stopping the lifting in step e, the method further comprises controlling the stacker to perform the action of lowering the bogie along the reverse direction of the lifting path by a certain distance.

5. The bogie stereoscopic warehouse management method according to claim 1, wherein The rule for determining in step e whether the monitored load current exceeds the reference current by a determined proportion is defined by the following formula: wherein, is the value of the monitored load current at the time point , is the value of the reference current at the time point , is a floating factor determined by the statistical distribution of current fluctuations of historical normal operation.

6. The bogie stereoscopic warehouse management method according to claim 1, wherein The method further comprises a verification step for confirming the validity of the successful access operation data, which is performed before step b, comprising: step f, obtaining a vibration signal through a vibration sensor arranged on the stacker during the physical interaction stage of the access operation; step g, extracting the peak amplitude and oscillation decay time of the vibration signal; step h, only when the extracted peak amplitude and oscillation decay time do not exceed the respective corresponding impact threshold calibrated by historical high-quality operation data, the operation is confirmed as a successful access operation, and the actual positioning coordinate corresponding to the operation is used for the judgment in step b.

7. The bogie stereoscopic warehouse management method according to claim 1, wherein The actual positioning coordinate in step a is a three-dimensional coordinate reading provided by a laser range finder or an encoder installed on the stacker at the moment when the physical action of the access operation is confirmed to be completed.

8. The bogie stereoscopic warehouse management method according to claim 1, wherein The method further comprises: when the stacker is in an idle state, controlling the stacker to move to one or more absolute reference points on the ground of the warehouse whose three-dimensional coordinates are constant for positioning measurement, and correcting the systematic measurement drift of the positioning sensor of the stacker itself according to the result of the positioning measurement.

9. The bogie stereoscopic warehouse management method according to claim 1, wherein The step c recalculates a dynamic physical offset based on the updated coordinate set, specifically: calculating the arithmetic mean of all actual positioning coordinates in the updated coordinate set on each coordinate axis in the three-dimensional coordinate system, taking the arithmetic mean as a new statistical center point; Then, calculating the vector difference between the new statistical center point and the theoretical design coordinates of the target storage location, taking the vector difference as the dynamic physical offset; The determination multiple is an integer greater than or equal to three selected according to the risk level assessment of the claimed method.

10. A bogie type warehouse management system characterized by comprising: The system comprises: A coordinate history database for storing actual positioning coordinates of multiple historical successful access operations for a target storage location in a warehouse; A control limit generation unit connected with the coordinate history database, which is configured to calculate and generate a numerical interval composed of a statistical center point and a determined multiple of the standard deviation of the coordinates based on the coordinates in the coordinate history database, taking the numerical interval as a dynamic control limit representing the statistical law of chronic evolution of the target storage location; A data acquisition interface for acquiring a single actual positioning coordinate corresponding to a new successful access operation after the operation is completed; A central processing unit connected with the coordinate history database and the data acquisition interface, which is configured to perform the following operations: judging whether the numerical value of the single actual positioning coordinate acquired by the data acquisition interface falls outside the dynamic control limit generated by the control limit generation unit; If the result is no, the single actual positioning coordinate is collected through the coordinate history database, and a dynamic physical offset is recalculated based on the updated coordinate set; If the result is yes, the single actual positioning coordinate is identified as a signal representing a structural acute mutation event, the coordinate is prevented from updating the dynamic physical offset, and an instruction for executing a locking treatment on the target storage location is generated; A motor current monitoring module for monitoring the load current of the stacker lifting motor in real time when the operation of taking goods from the target storage location is performed; And the central processing unit is further configured to: when the load current monitored by the motor current monitoring module exceeds the baseline current of the motor when it is unloaded under the same working condition by a determined proportion, it is determined that there is abnormal mechanical resistance and an instruction for stopping lifting is generated; A device control executor connected with the central processing unit for receiving and executing the locking treatment instruction and the stop lifting instruction.

Images